Web Browser Plug-in to Detect Fake Reviews in E-commerce Sites

Abstract

With the recent proliferation of online shopping customer usually publish reviews to the store and product after online shopping. At the same time, the online reviews become an important approach for the potential customer to know about the store and product. They usually check the online reviews to make decision whether to buy the product or not. Meanwhile, sellers and manufacturers are carrying out investigation of online reviews for decision making. Positive opinions can result in significant financial gains and/or fames for organizations and individuals. But to affect customers’ buying decisions, fake opinions are generated for purpose to promote special targets and/or denounce their competitors. Filtering out of untruthful information becomes an important issue in current situation. Throughout the period of project, I have analysed this issue and was able to introduced an enhanced method to identify the untrusted reviews as well as untrusted sellers and customers. I have introduced a web browser plug-in with an API which is capable of extracting the data (customer reviews, seller details, customer details) from e-commerce website and analysed the data using semi-supervised learning. After evaluating the results of the extracted comparing the already evaluated data set from the amazon website if shows the 80% of accuracy.